differing government and donor perspectives on …erd.eui.eu/media/backgroundpapers/mccord-differing...

Mobilising EuropEan rEsEarch For DEvElopMEnt policiEs

Differing government anD Donor perspectives on cash transfer baseD social protection in sub-saharan africa: the implications for eu social protection programming.

Anna McCord, Overseas Development Institute

paper prepared for the conference “promoting resilience through social protection in sub-saharan africa”, organised by the european report of Development in Dakar, senegal, 28-30 June, 2010.

2

AbstrActWhile donors are keen to promote cash transfer-based social protection in sub-Saharan Africa, and have promoted and financed a significant number of pilots to this end, such donor-led initiatives have not necessarily resulted in sustained or nationally-owned systems. Where cash-transfer programmes have been successful in the region, they have often been initiated domestically, rather than externally, and many with only limited donor support. This paper reviews recent experiences and draws on a set of commissioned studies exploring cash-transfer programming in Low Income Countries (LICs), to explore the possible reasons for differing government and donor perspectives on the desirability of cash-transfer programming, and attempts to abstract some broad policy insights. The paper highlights issues of national fiscal sustainability, government preferences, and ownership, as the key determinants of large scale government cash transfer implementation. The paper then draws out key lessons for future donor policy and programming in the region.

anna mccord Overseas Development Institute

[email protected]

The views expressed in this paper are those of the authors, and should not be taken to be the views of the European Report on Development, of the European Commission or of the European Union Member States.

Differing government anD Donor perspectives on cash transfer baseD social protection in sub-saharan africa:

the implications for eu social protection programming.

DIFFERING GOVERNMENT AND DONOR PERSPECTIVES ON CASH TRANSFER BASED SOCIAL PROTECTION IN

SUB-SAHARAN AFRICA: THE IMPLICATIONS FOR EU SOCIAL PROTECTION PROGRAMMING

3

List of Acronyms

AU African Union

CT Cash Transfer

CCT Conditional Cash Transfer

DFID UK Department for International Development

GBS General Budget Support

GDP Gross Domestic Product

GTZ Gesellschaft für Technische Zusammenarbeit

AIDS Acquired Immune Deficiency Syndrome

HSNP Hunger Safety Nets Programme

ILO International Labour Organisation LEAP

INGO International Non-Government Agency

ISSA International Social Security Association

LEAP Livelihood Empowerment Against Poverty

LICs Low Income Countries

ODI Overseas Development Institute

PEPFAR President’s Emergency Plan for AIDS Relief

PSA Programa de Subsidio de Alimentos

PSNP Productive Safety Nets Programme

PWP Public Works Programme Hunger Safety Nets Programme

SADC Southern Africa Development Community

SDC Swiss Agency for Development and Cooperation

SPF Social Policy Framework

SPF-I UN Social Protection Floor Initiative

UNICEF United Nations Children’s Fund

VUP Vision 2020 Umurenge Programme



4

Introduction

While donors are keen to promote cash transfer-based social protection in sub-Saharan Africa, and have promoted and financed a significant number of pilots to this end, such donor-led initiatives have not necessarily resulted in sustained or nationally-owned systems. Where cash transfer programmes have been successful in the region, they have often been initiated domestically, rather than externally, and many are with only limited donor support. This paper reviews recent experiences together with findings from two recently completed three year ODI research projects into cash-transfer programming and targeting,1 and draws on a set of commissioned studies which explore the political economy of cash transfer programming in Low Income Countries (LICs),2 in order to explore the possible reasons for the differing government and donor perspectives with regard to the desirability of cash-transfer programming, and attempts to abstract some broad policy insights. The paper highlights issues of national fiscal sustainability, government preferences, and ownership, as the key determinants of large scale government cash-transfer implementation. The paper then draws out key lessons for future donor policy and programming in the region.

Current cash transfer programming

Overview

There has been a growing interest in, and commitment to, the provision of social protection in sub-Saharan Africa over the last decade, among donors, governments and regional bodies (both the Southern Africa Development Community (SADC) and the African Union (AU), and a simultaneous increase in interest in the potential benefits of cash-transfer programming. Also, during the last decade, there has been a recognition of the need to move away from repeated humanitarian interventions in many parts of Africa, and to shift to the implementation of regular and predictable social protection programmes in order to address poverty in a way which can also assist in the mitigation of the impact of predictable disasters, and this has played a significant role in the development of social protection programming in general, and cash-transfer programming throughout the region.

The positive lessons emerging from Latin America have also stimulated increased interest in the potential role of cash-transfers internationally, and it has been estimated that, currently, there are more than 40 national cash-transfer programmes being implemented globally (Hanlon, Barrientos and Hulme, 2010). There is enthusiasm for such programmes among many donors, who perceive cash transfers as offering a cost effective and pragmatic means of delivering resources to the poor, and this has resulted in donor attempts to promote an expansion of cash transfer-based social protection throughout Africa. Interestingly, however, the attitude to cash transfers has been varied among governments in sub-Saharan Africa, resulting in programming on a notably smaller scale than in Latin America,

1 These research programmes were financed by the Swiss Agency for Development Cooperation (SDC) and UK DFID respectively.

2 The case studies were also part of the three-year study on cash transfers by ODI, funded by the Swiss Agency for Development Cooperation, and prepared in 2009 by Chinsinga, Ikiara, and Habasonda.



5

and less widespread, with many governments signalling ambivalence about this instrument, and with many donor-led cash-transfer pilots not being adopted by governments or implemented to scale.

Government interest in social protection and cash

transfers

There have been significant advances in the Social Protection discourse in sub-Saharan Africa in recent years, with the draft African Union Social Policy Framework (SPF) being presented in 2005 (and approved by the Heads of State in 2009), followed, in 2006, by the African Union Intergovernmental Regional Conference on Basic Social Protection in Africa, which took place in Zambia, which resulted in the Livingstone Call For Action on the Ddevelopment and Iimplementation of Nnational Ssocial Pprotection Pprovision. The year 2008 saw the first Regional Social Security Forum for Africa, held in Kigali by the International Social Security Association (ISSA), as well as the launch of the Africa Civil Society Platform for Social Protection, the first African Union Regional Consultations on Social Protection (in Kampala, Cairo and Senegal), and the first African Union Conference of Ministers in charge of Social Development, held in Windhoek, at which the Social Policy Framework for Africa was presented.

These initiatives have stimulated the development and adoption of a number of National Social Protection Strategies throughout the region, many of which include a statement of intent relating to the provision of basic social assistance through cash-transfer programming to complement a provision which has, to date, been largely restricted to contributory social pension provision for state employees, following systems often established during the colonial period. While, in some instances, there has been a commensurate development or expansion of nationally-initiated social protection programming, including cash-transfer development, particularly in the Southern Africa region where six countries are implementing national cash-transfer programmes. Three of these programmes were developed independently of donor input (technical, administrative or financial), while elsewhere, donors (both bi- and multi-lateral agencies and INGOs) have sought to stimulate the development of programming in order to realise these strategies, often by financing the implementation of cash-transfer pilots.

While there have been some notable successes in donor-supported cash transfer programming (most notably in the case of the cash components are the Productive Safety Nets Programme (PSNP) in Ethiopia, and, on a smaller scale, the Rwandan Vision 2020 Umurenge Programme (VUP)), many donor-supported cash-transfer pilot initiatives have not resulted in the development of nationally-owned or nationally-financed cash-transfer programmes which are progressively going to scale. This paper explores the reasons for this, and what we can learn from the limited success of pilot-based cash-transfer programming initiatives.

Donors and cash transfers

Cash transfers are popular with many donors and development academics (Hanlon et al., op cit). This is, in part, due to the well-documented success of conditional cash-transfer programmes (CCT) in Latin America, such as the Oportunidades programme in Mexico (formerly known as Progressa), Chile Solidario and the Bolsa Família in Brazil, which are large-scale national programmes which reach a significant percentage of the poor, at costs which are generally considered fiscally sustainable, at 0.4% of GDP for Brazil and Mexico, and 0.1% for Chile



6

(Bastagli, 2009).3 Conditional cash-transfer programmes, which are conditioned upon the basis of service utilisation, are not easily transferable to the sub-Saharan context in the absence of widespread service-provision, and this has been generally recognised in the types of cash-transfer programmes which have been promoted by donors, and, to a lesser extent, by governments in the region, which have not, for the most part, attempted to include conditionalities based upon the usage of health or education services.

The ILO has identified a basic package of the cash transfers and the social provision required to address the minimum needs of the poorest, a package which includes cash transfers for children, the elderly and the disabled, for public-works employment for the working-age poor, and basic health and education provisions. Drawing on a number of country case studies in the region, the ILO estimated that the mean cost of the provision of the cash transfer and public-works component of the basic social-floor package (excluding service provision) in sub-Saharan Africa would represent, on average, approximately 4.4% GDP (ILO 2008). On this basis, it has been argued that such provision is, in the medium term at least, potentially domestically-affordable, subject to initial donor support. This package was launched in April 2009, with the endorsement of the wider UN community, (and over 190 governments) as the UN Social Protection Floor Initiative (SPF-I), highlighting the current preference for cash transfer-based social protection, and recognising the need for the harmonisation of donor programming in this sector.

In an effort to promote the adoption of social protection in general, and targeted cash-transfers in particular, the donor community (bi-lateral and multi-lateral donors, INGOs and UN agencies) have engaged in a range of activities, including supporting and financing the development of a civil society movement to call for protection provision (the Africa Civil Society Platform for Social Protection, and its regional contributors), supporting the AU Social Policy process, financing training of regional and national civil servants and politicians, contributing to large-scale social-protection programmes which include cash-transfer components (the VUP in Rwanda, the PSNP in Ethiopia, and the PSA in Mozambique), and financing numerous pilot programmes to demonstrate the effectiveness of cash transfers to national governments in contexts in which such programmes are not currently preferred policy instruments. The most celebrated cash-transfer pilots in the region in recent years have been the “Kalomo” pilot in Zambia, financed by largely by GTZ, Irish Aid and DFID, and the UNICEF supported “Mchingi” programme in Malawi, which has diversified into a set of five pilots supported by a range of different agencies. Such pilots have been successful in terms of addressing poverty among eligible groups within their target area, but have not typically been adopted by national governments or taken to scale, and have been criticized for:

“Creating temporary islands of access to internationally financed social welfare.” (Devereux and White, 2010:73)

3 It is important to note, however, that Bastagli (2010) highlights the fact that, even in the Latin American context, in many countries (excluding Brazil, Mexico and Chile) it is hard to make the case that CCT are “fiscally sustainable” since many are primarily externally-funded, and therefore tend to be short lived, despite recording impressive impacts on intended outcomes; see, for example, the case of Nicaragua, highlighted in Bastagli, 2009. Bastagli also offers the critical insight that the costs of these transfers, the official estimates reported of costs as share of GDP, typically do not include the full administrative costs associated with programme implementation (Bastagli, 2010).



7

The limited success of these programmes is critical, as it questions the efficacy of an approach which has been dominant in the donor community in recent years, and has absorbed considerable donor finance, as well as national and local resources in participating countries, highlighting the need for self-reflection among the donor community. This paper attempts to explore some of the reasons for the reluctance of some African governments to adopt donor-piloted models of cash-transfer provision, and to assess the implications for future collaboration and donor activity in this area.

Current cash transfer programming

Before examining the limited success of the donor-supported pilot cash-transfer programme model, it is important to acknowledge that significant large-scale cash-transfer programming is currently taking place in sub-Saharan Africa. There is considerable variation in cash transfer programming in the region, and the main patterns of programming are illustrated by the fourteen countries outlined in Table 1 below.



8

Table 1: Cash Transfers and social protection programming in selected SSA

countries

National Cash

Transfer

Programme(s)

Type of CT Cash Transfer Pilots

(Major international

sponsor)

Other Major

Intervention

South Africa Y Pension and

others

PWP

Botswana Y Pension

Namibia Y Pension Y

Lesotho Y Pension Y

Mozambique Y Pension*

Swaziland Y Pension Y

Ethiopia Y Complement to

PWP for labour

constrained

PWP

(Rwanda)** Y Complement to

PWP for labour

constrained

Y PWP

Malawi Y Input Subsidy

Programme

Zambia Y Targeted Food

Security Pack,

Fertilizer

Support

Programme

Ghana Y

Kenya Y

Tanzania Y

Uganda Y

* The Mozambican PSA is not targeted exclusively to the elderly, but primarily reaches this group. ** The Rwandan VUP is a national programme, which forms a central component of Rwanda's Economic

Development and Poverty Reduction Strategy (EDPRS), but is not yet nationally implemented.

While there is significant diversity within these countries, they can be used to illustrate of four broad scenarios, representing differing approaches to cash transfer programming. National cash transfer programmes are implemented in eight of the fourteen countries; South Africa, Botswana, Namibia, Lesotho, Swaziland, Mozambique, Rwanda and Ethiopia. The first grouping comprises the first six programmes in the table, these have national coverage and are implemented and financed by the government which also developed and initiated the programmes. The Mozambican PSA is included in this group, as, while it was expanded with international donor support in recent years, it was domestically-initiated and financed prior to this. In all instances, with the exception of South Africa, the cash transfer takes the form of a pension, while, in South Africa, a comprehensive range of transfers is provided, reaching 12 million people out of a population of 44 million, in line with the Social Floor package (targeting children, the elderly and the disabled, among others).



9

The second grouping of countries, (Ethiopia, and, to a lesser extent, although following a similar the same model, Rwanda), also implement nationally co-ordinated cash transfer programmes). However, they differ from the first group, in that both programmes received significant donor inputs during programme development, and are reliant on ongoing finance from external donors for their continued operation. In these countries, the cash-transfer programmes do not take the form of pensions, but are designed as complementary transfer programmes to accommodate labour-constrained households whose members are not able to participate in the national Public Works Programmes which form the major component of national social-protection provision.

Malawi and Zambia, the third grouping, enjoy a multiplicity of donor-financed cash-transfer pilots with varying administrative, financing, and targeting and design features, which have featured prominently in the international social-protection discourse in recent years, especially the “Kalomo” and “Mchinji” pilots, whose successes in terms of impact, have been widely communicated within the region. However, despite these pilots having been implemented since 2003 and 2006 respectively, the government of Malawi and Zambia have not allocated significant domestic resources for the implementation of these pilots or their expansion into national programmes, and the major national state-sponsored social-protection interventions are the Input Subsidy Programme in Malawi, and the Targeted Food Security Pack, and the Fertilizer Support Programme respectively, which do not conform to the conventional portfolio of social-protection instruments prescribed by the international donor community. In recent months, subsequent to the provision of a significantly increased ten-year financing-package from the UK DFID, the government of Zambia has, however, increased its own domestic commitment to cash-transfer programming, an issue which will be discussed below.

In the fourth grouping, comprising Ghana, Kenya, Tanzania and Uganda, a range of donor-supported cash-transfer pilots have also been initiated, with varying levels of government engagement, but have not yet gone to scale In Uganda and Tanzania, there is very limited government social-protection provision, and, despite donor-financed pilots, there is little current evidence of national cash-transfer programme development on the basis of these pilots. In Ghana, while the celebrated donor-supported cash-transfer pilot Livelihood Empowerment Against Poverty pilot, better known as LEAP, enjoys government support, and represents a core component of the government social-protection strategy, it remains limited in coverage, and in Kenya, while two donor-financed and initiated cash-transfer programmes aimed at children and those in the drought-affected north of the country respectively, are currently being piloted, a government initiated and financed pension-pilot is simultaneously being rolled out, with the objective of attaining national reach. The simultaneous roll out of three cash transfer programmes, adopting differing criteria and implementation modalities suggests potential efficiency losses in a resource-constrained environment, and raises questions regarding the future sustainability, and the degree of government buy-in to the two non-government initiated programmes.

This brief overview illustrates a diversity of national responses to cash transfer programming, and suggests that governments may have a preference for nationally-initiated programmes in general, and specifically those which target either i) the elderly, or ii) the residual labour-constrained poor, unable to participate in a PWP-based social-protection system, or provide an alternative form of protection for the poor, for example through productivity enhancing interventions based on agricultural inputs, which, while “socially protecting”, do not conform to core social protection programming, as commonly conceptualised in donor thinking.



10

This profile represents a challenge to donor perspectives on cash transfer programming; while donors and NGOs are keen to promote the concept of cash transfers and to demonstrate their success at achieving the transfer of resources to the intended target groups through multiple pilots around the continent, the limited success apparent from the overview, in terms of national “take up” above, suggests that there may be a problem in terms of the traction of donor-led pilots, and a difference between government and donor perceptions that cash transfer programming, as conceptualised by donors, represents the optimal social protection response.

It is instructive to examine why this may be the case, and what the donor and research community can learn from this. These questions will be explored using evidence from a three year ODI research programme into cash transfers, and case study research commissioned into national perspectives on conditional cash transfer programming in three low income sub-Saharan African countries; Kenya, Malawi and Zambia.4

Cash transfer programming in three Low Income sub-

Saharan African countries

Drawing on the three case study countries of Kenya, Malawi and Zambia, data on cash transfer cost, coverage and domestic financing allocations are examined, and national perspectives on the adoption of cash transfer programming to provide social protection provision are explored in an attempt to understand the factors which inhibit the take up of cash transfer programming. First, the coverage of current cash transfer programming in all three countries is reviewed in order to assess the current performance of the donor pilots, and then current and projected national government allocations to these programmes are reviewed on the basis of their medium-term expenditure-frameworks or similar, as an indicator of the level of national commitment to these programmes, and, finally, the fiscal implications of programme extension to cover all eligible beneficiaries (going to scale) are estimated, before conclusions are drawn with regard to national policy preferences.

Cash Transfer Coverage

There are a range of social protection initiatives in each of the three case study countries, including a number of cash transfer programmes, all of which are currently in pilot or initial roll-out stages, with none being implemented on a national scale. All but one of the programmes (the Kenyan pension scheme) were initiated by the donor community. Kenya has three main cash-transfer pilots: i) the programme for Orphans and Vulnerable Children (OVCs), initiated by UNICEF and with multiple donors, ii) the Hunger Safety Nets Programme (HSNP) for those living in destitution in arid areas, implemented with UK DFID support, and iii) the government initiated and financed national social-pension. Malawi has five differing pilot components linked to the “Mchinji” initiative, and Zambia has only one major cash-transfer pilot programme. All of these programmes focus primarily on transfers to “ultra-poor” households, crudely conceptualised as the poorest 10% of the population. An assessment of coverage of these programmes in 2009 indicated that, when considered at national level, the coverage of these pilot programmes in terms of the

4 The case studies were produced by Ikiara, Chinsinga and Habasonda respectively.



11

poorest households considered eligible5 is extremely low, ranging from 3% in Zambia, to 8% in Malawi, and 9% in Kenya (if coverage of all three pilot programmes is aggregated); see Table 2.

Table 2: Summary of Main Cash Transfer Programmes and Coverage, in

Kenya, Malawi and Zambia

Source: McCord (2009), based on Chinsinga (2009), Habasonda (2009) and Ikiara (2009)

Such low coverage should be anticipated given these programmes’ pilot nature, but it is important to be aware of the actual levels of coverage, when the ‘success’ of donor pilots is being discussed in the international discourse, as this provides an empirical ‘reality check’ on the scale at which such programmes are operational, and the scale of expansion which implementation on a national scale would represent, and also that these programmes are operating with significant, and, for the most part, un-costed levels of donor administrative support. It is also important to note that these coverage figures represent only the level of coverage among the poorest households. If calculated in terms of all poor households, these programmes currently cover less than 1% of all poor households in Zambia, 2% in Malawi, and 4% in Kenya, and if fully implemented cover only 15%, 23% and 12% of all poor households respectively, see Table 1.

It is important to highlight the low coverage of these programmes, particularly those which are donor-led, as they are often presented in the international discourse as though they represented significant (large scale) responses to poverty, rather than small scale pilots with latent, rather than patent, potential for significant national social protection impact.

If implemented nationally, these programmes would exclude more than one million poor households in Malawi and Zambia, and more than two million in Kenya

5 The eligible group of poorest households is variously described in these countries as the ‘ultra poor’, ‘extreme poor’, ‘hardcore poor’, ‘incapacitated’ or ‘non viable’ households, which are estimated to comprise 10% of the overall population in Malawi and Zambia, and 19% in Kenya).



12

(Table 1), given the strict rationing of access, as the targeting criteria restrict eligibility to the lowest decile (or, in the case of Kenya, the lowest quintile) of the population, and, in some instances, deliberately exclude households with members of working age and low-dependency ratios (Zambia).

The exclusion of households that are not labour constrained is problematic, as it is based on the assumption that all labour has the potential to be used productively, and to generate an adequate income, an assumption which is widely-held among donors and governments in sub-Saharan Africa. This fails to recognise the current nature of labour markets in many low income African countries, in terms of the structural nature of under-employment and unemployment, and the phenomenon of the working poor (Slater and Farrington, 2009. Wood, 1999 and du Toit, 2004), excluding these groups from the reach of cash-transfer provision. In some instances, such groups are accommodated through public works programming (PWP) (for example, in Malawi) but the numbers of the working-age poor tend to greatly exceed available PWP employment opportunities, resulting, in most cases, in strict rationing of access to temporary PWP employment, and low levels of coverage among this group. Only in the case of Ethiopia is PWP employment provided on a mass scale, offering a form of employment guarantee.

Government allocations

In order to assess governments’ own level of engagement with these cash transfer pilots, it is instructive to examine the scale of government allocations to these programmes from domestic funds, which varies considerably.

In 2009, the Zambian government contributed approximately 5% of the budget of the cash transfer pilot (equivalent to approximately 1% of the resources allocated by the government to the state employee pension scheme). In Malawi, the government does not anticipate making any financial contribution to the cash transfer pilot until 2011-2012, and will then contribute only 12% of the total cost per annum as the programme is extended (Chinsinga, 2009). These low levels of domestic contribution imply that government ‘support’ for cash transfer programming may not be matched by a commensurate domestic fiscal commitment, and that programme commitment at the level of the relevant ministry, may not be reflected in a similar level of commitment at the level of Ministry of Finance. The cash transfer pilots in Malawi remain dependent on continued and increasing donor funding both for their continued implementation and for any anticipated expansion, given the absence of a medium-term domestic financing plan. This reliance on ongoing external donor inputs implies that cash transfer programming may not be a national policy priority and that the pilots are unlikely to become domestically championed components of a national social protection strategy. Developments in the mid 2010s in Zambia, however, where UK DFID has introduced an extended medium term financing commitment to underwrite the extension of the pilot with guaranteed financing for a ten year period, have stimulated a shift in the government financing response, and resulted in the development of a medium term financing plan on the part of the government, wherein government financing increases incrementally to cover the majority of programme costs by the end of the donor financing period. This suggests that instruments to promote medium term donor financing credibility may have an impact on government willingness to take on the future liabilities implied in adopting and extending cash transfer pilots.

The Kenyan situation differs somewhat, with the government in 2009 contributing 30% of the costs of the donor initiated child benefit programme, although planned contributions will reach a ceiling of 20% in the coming years as



13

coverage increases, implying a significant continued reliance on donor inputs for the support and extension of this programme in the medium term (Ikiara, 2009). In contrast, the cash transfer pilot for the elderly, which was domestically initiated, is financed wholly from domestic resources and has a domestic financing plan for the roll-out to national coverage. This reflects Kenya’s concern to promote policy-stability and to safeguard the predictability of resource flows to priority groups by limiting reliance on donor funding, following their experience of the suspension of General Budget Support in the wake of the political crisis of 2008. In terms of real financing-flows, of the three cash-transfer programmes, the greatest proportion of domestic resources goes to the new non-contributory pension scheme, and the lowest proportion to areas with significant and ongoing donor-funding, such as the drought-affected Sahel area covered by the Hunger Safety Net (HSN) programme, and the child-benefit programme, which has received significant PEPFAR and Global Fund financing. The level of donor reliance for the financing of these projects may potentially have implications for future programme sustainability, in the absence of long term donor financing commitments, although emergency related programmes, such as the HSN (and PSNP in Ethiopia) tend to be more assured of repeated donor financing, which is somewhat insulated from policy preference shifts in development financing, even in the absence of explicit medium term financing agreements.

Implications for programme sustainability

If fiscal allocations to cash transfer programming are taken as an indication of state commitment to social protection provision, national ownership of the pilot programmes under discussion is limited, which suggests that the implications for programme scale up and future sustainability may also be limited.

Given the low level of domestic financing for existing pilot provision, and the limited commitments to future programme extension, with the exception of the pension in Kenya, and the Zambia cash transfer programme, the other programmes are only affordable, even as pilots, on the basis of continued donor provision, and the fiscal implications of taking these programmes to scale is discussed below.

This analysis of domestic allocations to cash transfer programming suggests that there may be a gap between the rhetorical endorsement of cash transfer provision made in response to donor financing incentives, and a domestic fiscal commitment to supporting such programmes, even in the case of high profile cash transfer programmes such as those in Zambia and Malawi, and that changes in donor financing credibility, may have an impact on shifting these commitments.

Taking these limited levels of national financing into account, the pertinent question becomes what factors are inhibiting government take up?

Potential factors inhibiting cash transfer

programming

Based on the findings of the ODI cash transfer research project, and the three country case studies into the political economy of cash transfer programming which informed the discussion above, six main factors can be identified as potentially limiting government engagement with cash transfer pilots in the region; fiscal concerns, preference for productive safety nets, concerns regarding dependency and labour market distortion, political economy considerations, donor inconsistencies, and donor targeting preferences.



14

i) Fiscal concerns

The case studies suggest that the limited engagement with cash transfer programming as envisaged in donor pilots is largely the result of binding fiscal constraints faced by low-income countries with low domestic resource bases, combined with a reluctance to carry out budgetary reprioritisation in favour of social protection provision, for reasons of alternative domestic policy preferences which have a prior claim on the fiscus, and political economy considerations, which will be discussed below.

Governments are also concerned with regard to the significant long-term fiscal liability that would be implied if cash transfer programmes were extended beyond a limited closely defined sub-section of the poor, particularly if access to such a programme were to be perceived as a right. In this context large scale social transfer programmes may be regarded as unaffordable, and represent a significant fiscal risk, in terms of future liabilities, particularly given the uncertainties associated with short term donor aid flows, and hence incentives to extend domestic liabilities to cover such programmes are weak, and disincentives abound.

In order to illustrate the issue of future fiscal liabilities and affordability in relation to the three case study countries, the cost of the coverage of all ultra-poor households, on the basis of the pilot programmes, are projected below. The cost of taking the programmes ‘to scale’, offering full coverage of the subsection of the poor population deemed eligible for CT receipt in pilot programme design (10% of households in Malawi and Zambia and 19% in Kenya), would be between 0.5 and 1.7% of GDP, or 2 to 4% of the total government budget in these three countries (Table 3).

Table 3: Indicative Cost of National Provision

Source: McCord (2009), based on Ikiara (2009), Chinsinga (2009) and Habasonda (2009)

It is interesting to note that these figures are relatively high compared to the cost of the major CCT implemented in Latin America, with the Bolsa Família which reaches 26% of the population in Brazil costing 0.4% of GDP, the Mexican Oportunidades 0.4%, and the Colombian Familias en Acción 0.1% of GDP (Bastagli, 2009).

However, these figures illustrate the cost of addressing the needs of only a small and arbitrarily defined segment of the poor on a national basis. They do not represent the cost of social protection coverage for all the poor, as the restrictive targeting they entail would result in low coverage and high levels of exclusion of the poor (see Ellis, 2008). Without allowing for possible economies of scale, extending



15

similar transfers to all the poor in the case study countries would require approximately 2 to 7% of GDP. Such coverage would represent a significant and potentially unsustainable drain on the fiscus if domestically financed.

Affordability

This illustrates the ‘affordability’ challenge faced by LICs in terms of cash transfer provision, even if only targeted to a subset of the poor, and may represent a challenges to the appropriateness of adopting Latin American models for cash transfer programming in Africa, given the primarily domestic funding base for cash transfer programming in Latin America and the middle-income context, and highlights the need for ongoing external financing if programmes are to be implemented at scale. It has been mooted by the IMF that the crisis may result in marginal increases in fiscal space in LICs,6 but there is no evidence that the social-protection sector would successfully compete against the claims of other more politically-powerful sectors on this space. Given other competing demands on the fiscus, and the inadequacy of government resources to cover existing commitments in a range of priority development sectors (health, education, sanitation, agriculture, infrastructure etc), low-income countries cannot afford national cash transfer programmes with broad eligibility criteria, funded from domestic resources, in the absence of significant reprioritisation of budgetary allocations. Cash transfer programming could be sustainable domestically with a significant shift in domestic budget priorities, which in turn is contingent on the strength of the political lobby for cash transfer programming, and government preferences for provision, but in the absence of such reprioritisation, the continuation of some of the cash transfer pilots in the case study countries is contingent on ongoing donor support.

Hence, for sustained programming to scale, ongoing external financing may be required. In the light of this, for reasons of fiscal prudence, given the short term nature of most donor financing for cash transfer programming, allied with the unpredictability and capriciousness of donor financing flows, governments may wisely be reluctant to commit resources to pilots or their extension in the short or medium term, in order to safeguard their public finance management credibility, and avoid setting up programming, and associated expectations, which may have to be terminated if donor financing priorities shift.

ii) Productive investment

In addition to the fact that cash transfer provision may imply significant future liabilities that cannot be underwritten domestically, and result in ongoing reliance on external funding sources, potentially rendering national governments vulnerable to future shifts in donor financing preferences, there is a second aspect to the reluctance to commit national resources to such programmes, a preference for the allocation of domestic resources to productive, rather than consumption, expenditure. Cash transfers are often considered as consumption rather than investment expenditure by governments – addressing individual poverty but not contributing to broader economic growth. Some governments, including those of Zambia and Malawi, have been reluctant to fund this form of social protection from limited domestic resources, and prefer to invest in ‘productive’ social protection interventions which fall outside the boundaries of conventionally defined ‘social

6 Gupta S, presentation at European Report on Development conference on ‘Experiences and Lessons from Social Protection Programmes across the Developing World: What Role for the EU? Paris, 17-18 June 2010.



16

protection’ such as agricultural input or subsidy programmes that promote production at household level.

Hence, governments have two linked concerns about the fiscal implications of cash-transfer programming; on the one hand, financing cash transfers represents non-productive consumption rather than investment expenditure, and, on the other, financing such an intervention risks committing themselves to potentially unsustainable recurrent budget expenditure. These concerns are particularly acute in low income countries where there is no existing long-term, on-budget social security system.

This indicates that medium to longer-term programme ‘sustainability’ is dependent on continuing donor allocations, and so much cash transfer provision in Malawi and Zambia, and to a lesser extent in Kenya, is ‘affordable’ and ‘sustainable’ only as long as donor support continues, and any extension of provision is contingent on increased donor allocations.

Ongoing reliance on donor inputs does not, however, necessarily imply a lack of programme sustainability; the PSNP in Ethiopia and, to a lesser extent, the PSA in Mozambique, are dependent on ongoing donor financing to meet the cost of national implementation, but, in these instances, donors have collectively made financing available on a medium term or recurrent basis, in partnership with the government to support national implementation. This situation differs significantly from situations in which a more limited number of donors are financing a small pilot in the short-term, without a commitment to finance the programme in the medium-term as coverage is expanded, i.e., where donors are explicitly financing the pilot as a pilot to demonstrate the concept, rather than making a commitment to share the cost of expansion in the medium term.

These programmes indicate that financing predictability, with guaranteed medium-term financing flows, and donor rationalisation, with coordinated financing, rather than financing based on individual donor preferences based on institutional mandates or instruments, which may be potentially ‘competing’, may be more appropriate to promote government engagement with cash transfer programming.

iii) Dependency

Governments throughout the region, and many donors articulate a concern lest concerned lest cash transfer programmes create dependency, welfare traps and distort the domestic labour market, and, for this reason, cash transfer programming in the region tends to exclude the working age poor, and focus on labour constrained households, leaving public works employment as the primary social protection option for the working-age poor, as noted above.

The concerns of policy makers and civil servants, particularly in Africa, about dependency and fiscal liabilities are often articulated around cash transfer programmes and restrict the adoption of cash transfer programming as a social protection instrument. Concerns about labour market distortion, in general, and dependency, in particular, make governments keen to restrict the scale of cash transfer provision. With the exception of some humanitarian interventions, such as the cash transfer programme for vulnerable drought-affected households in Kenya (Hunger Safety Nets Programme) funded by UK DFID, the poor of working-age tend to be excluded from eligibility for cash transfer programmes. Instead, governments focus, for example, on households with vulnerable children (Malawi and Kenya), the poor with limited labour (Zambia), or the elderly (Kenya). This has led to a reworking of the concept of the ‘deserving poor’ in recent years, based on the assumption that households with available labour have the potential to generate



17

income, regardless of labour market realities. Underlying this position is an assumption that labour markets function adequately, with sufficiently remunerated employment available for all those seeking it. It does not take into account the chronic levels of unemployment, under-employment and the existence of the working poor or adversely incorporated – those who are in the labour market, but receive incomes so low that they remain below the poverty line (Wood, 1999) – situations that are common in most low-income countries, and limit the actual, rather than potential, productive capacity of household labour.

Despite this reality, labour availability is used explicitly as a criterion for exclusion from cash transfer programme participation in many cases (for example in Malawi and Zambia). Even where extreme poverty is the explicit criterion for programme participation, households of equal poverty but with available labour tend to be excluded, with labour availability serving as a proxy criterion for programme exclusion, reflecting this dominant concern regarding dependency. This concern serves as an additional incentive to favour directly ‘productive’ over consumption based forms of social protection. This concern also results in a widespread preference for the implementation of public works based social protection over cash transfers, as the work requirement directly addresses concerns regarding dependency, and also provides a form of social protection which is perceived as inherently ‘productive’.

iv) Political Economy Considerations

Political economy considerations are also key determinants of government perspectives on cash transfers, and budgetary allocations to cash transfers do not simply reflect absolute fiscal constraints or the availability of external resources, as engagement with cash transfer programming is an issue of policy choice, susceptible to political preferences and constituencies.

Research in Kenya (Ikiara, 2009) indicates that state ownership is not just a consequence of the cost and fiscal implications of cash transfer programmes, but also informed by the challenges facing a state at any particular time, and the role of cash transfers in addressing these challenges. Cash transfers can play a role in the legitimisation of the state and the promotion of the state-citizen compact. In Kenya, for example, the government is extending cash transfer provision, and making significant fiscal allocations to social protection, even in the context of the financial crisis, in an attempt to promote stability following the civil disturbances of 2008. Similarly the implementation of cash transfer programmes can be used to extend support to populations with limited allegiance to the state, and, at the same time, establish a symbolic legitimacy for a conflict government, in terms of its ability to honour the state-citizen compact in a post-conflict context. Hence, it is important to recognise that cash transfers can have an instrumental role in relation to stabilisation (and state formation) in fragile states, in addition to their direct function in terms of the reduction of poverty and vulnerability.

The level of government commitment to cash transfer programming, and the extent of fiscal re-allocation to support such programmes, is thus, in part, contingent on whether a government needs, for stability and political survival, to ensure that cash transfer programming reaches certain groups. In such instances, it is more likely that a state will allocate significant domestic resources to cash transfers, as in Kenya, than in more stable contexts in which a cash transfer is likely to be of more limited significance in terms of stability. Even where a government is unable to contribute significant financial resources, its level of political support for cash transfer programming is likely to be greater if successful programme implementation will have beneficial political outcomes. Hence countries are more likely to support social protection interventions actively and allocate domestic resources to cash transfers where there is either i) explicit popular demand for provision or ii) a



18

perceived political need for the government to transfer resources to the poor to promote national stability or retain power, as in the case of many post conflict states, or Kenya in the wake of the civil unrest of 2008, and iii) where governments do not want cash-transfer programming to be vulnerable to the vagaries of donor policy preferences and funding fluctuations.

Popular movements demanding cash transfer provision

It is interesting to note that in neither Zambia nor Malawi is there an organised mass popular lobby for the provision of social protection in general, or cash transfers in particular, and such policies are not explicitly linked to a significant electoral constituency, unlike the situation in Latin America and parts of South Asia (most notably in the Indian context). This is symptomatic of the region overall, where there is a lack of collective action around the poverty agenda and the poor have not asserted themselves in the political arena, and donor attempts to stimulate a civil society movement in support of social protection provision have borne limited fruit. For these reasons social protection provision is not, in many instances, an electoral issue, and incentives for policy development in this area remain limited. However, where there is a latent or patent expectation of some form of natural resource dividend which is recognised by the state, and some form of re-distributive intervention is anticipated by the population, an incentive for cash-transfer programming is more apparent, which may be related to the engagement of the Ghanaian government in LEAP, and their concern to develop complementary interventions to promote coverage, as the institutional constraints to the LEAP extension become increasingly apparent.

Stability

Where social stability is a concern, for example, in fragile or post-conflict contexts, in situations of latent or patent civil unrest, or instances where there is an expectation of some form of peace dividend, or resource dividend, (as with the Ghanaian oil dividend), there is a greater incentive for governments to adopt cash transfer programming, although government preferences may be for productive social protection (such as public works) or agricultural programming, in preference to cash transfers. In some cases, however, the symbolic nature of a cash transfer may serve a political function in fragile situations, and be more readily implementable, in terms of the rapidity of implementation and administrative feasibility.

Where political economy considerations do lead countries to favour the provision of cash transfers, as in Kenya, experience of donor policy shifts and changes in financing preferences may result in a reluctance to place such a significant redistributive policy at risk through ongoing donor reliance, and may promote increased domestic financing of sensitive cash transfer programmes, and simultaneously increase reluctance to implement large scale donor dependent programmes.

v) Donor inconsistencies and programming preferences

Given the substantial influence over the social protection policy process which donors wield due to their financial and technical resources, the lack of donor harmonisation at country level is a key challenge, with donors competing to promote their own programming preferences, in terms of target groups, design, and implementation modalities, often informed by institutional mandates and priorities, rather than the need for coherent or government led programming.

There is a need to avoid competition between donors over differing cash transfer models, the implementation of multiple donor initiated programmes



19

simultaneously, and the provision of inconsistent advice and conflicting advocacy messages, between which governments may not be well placed to mediate, particularly given the power imbalance between donors and governments, and the diplomatic imperative of maintaining positive relationships with donors proposing divergent policy and programming activities. To this end, there is a need to rationalise programming and financing in relation to cash transfers, and to minimise donor competition around alternative approaches and instruments, and to address cash transfer provisioning from a perspective which takes government preferences into account, and does not fragment provision and programming.

vi) Donor Targeting Preferences

In addition to the factors identified above donor targeting preferences may also reduce the attractiveness of some donor designed cash transfer pilots to national governments. While nationally initiated programmes in the region have focused on the provision of cash transfers for a tightly defined and easily identifiable group of beneficiaries, based on categorical targeting, primarily the elderly, donors have promoted a range of targeting criteria, which may not be so readily acceptable to governments concerned about containing future liabilities, programme scale, and ease of implementation.

Donors have tended to advocate for support for various different target groups according to their institutional mandates and programming preferences, with popular target groups being the elderly, children, and “the poorest 10%”. While targeting the elderly implies directing resources to a readily identifiable, socially acceptable, and limited group within the population, targeting children opens the state to a far larger group of potential claimants and increased fiscal liabilities. Similarly, targeting the “poorest” may not be easily containable, in terms of the potential claimant base, and also may be less socially acceptable, and may fail to conform to the prevailing views of poverty, with numerous anecdotal reports of beneficiaries rejecting the sub-division of the poor into such subtle categories and sharing the available/0 benefits in order to increase the equity of the transfer within a community in defiance of donor directives. Stewart and Handa suggest that targeting exclusively to the lowest decile is based on donor orthodoxies, rather than any empirical rationale (Stewart and Handa, 2008). These concerns are borne out empirically in the work of Ellis, who illustrates the marginal differences in poverty between the bottom five deciles in many low income sub-Saharan African countries, and highlights the inequality of prioritising the needs of one decile over another in cash-transfer programming (Ellis, 2008).

Lessons from the Productive Safety Net Project (PSNP) in Ethiopia suggest that social divisions at community level between programme beneficiaries and non-beneficiaries may emerge when the poor are numerous within a given community but programme resources and targeting criteria mean that only some receive benefits. There a risk that attempts to target the very poorest households may result in the exclusion of those who are only marginally less poor with the consequence that in terms of income and welfare, beneficiaries ‘leapfrog’ over those who are only marginally richer (ibid).

Donor favoured means-tested approaches may also be problematic in many income sub-Saharan African contexts. Approaches which make use of a poverty threshold (measured directly or through proxies) require significant human and financial capital, as they need careful monitoring, frequent updating and pose complex problems of interpretation, and hence discretion for programme administrators, thereby increasing cost and also potential for corruption. One way to resolve these trade-offs is by using other targeting approaches, such as those identifying the poor based on geographical criteria or according to social or



20

demographic category (such as the elderly, orphans and vulnerable children, women-headed households etc). Those approaches based on social categories or geography may have a stronger appeal to the governments of poor countries, not least because their implementation is low-cost relative to that of approaches based on income (Slater and Farrington, 2009), although the poverty targeting of such approaches may be significantly less effective, resulting in greater exclusion of the poor than poverty-focused targeting (Handa 2008, Slater and Farrington, op cit).

It is interesting to note that while the targeting of children has not emerged as a priority from within nationally-financed programmes, the existence of relatively plentiful external resources for HIV/AIDS related programming through PEPFAR and the Global Fund during the 2000s, has meant that significant resources have been available to finance cash transfer pilots addressing the needs of orphans and other vulnerable children. This has influenced the nature of cash transfer programming in the region, generating a demographic focus on children and colouring the development of the cash transfer discourse, which may not necessarily be consistent with domestic priorities in terms of social protection provision, leading to the exclusion of many of the poor from cash transfer programming, and also potentially reducing the likelihood of national programme adoption based on domestic financing.

Hence donor targeting preferences may also serve to inform the reluctance of governments to buy in to pilot models, which are externally determined, and may not conform to their own priority targeting preferences.

Inhibiting Factors

Six main issues have been outlined above as potential inhibiting factors which serve to constrain government engagement with donor led cash transfer piloting in the region; fiscal concerns, preference for productive safety nets, concerns regarding dependency, political economy considerations, donor inconsistencies and mandate based approaches, and donor targeting preferences. These issues serve to challenge prospects for government engagement and ownership, and would merit consideration in future programming. These insights may offer lessons for the development of EU policy and programming in the social protection sector, as the EU has the potential to contribute to a resolution of these issues, not only through its own programming, but also by encouraging improved co-ordination on the part of the Member States and the activities of other agencies.

Donor Influences on Graduation and Growth

Expectations

In addition to the factors discussed above, which may be inhibiting national government ‘take-up’ of cash transfer programming, there is one additional factor relevant to consider in relation to donor advocacy around cash transfer programming.

In addition to determining the instrument and the target group, donors influence programme design and the cash transfer discourse in other significant ways. Given many governments’ articulated concerns to promote productive safety nets, and avoid long term transfer liabilities, donors have stressed the linkage between cash transfer programming, national growth and individual graduation in the regional discourse. It can be argued that these aspects of cash transfer programming have been overstated by donors in order to encourage national buy-in and have come to shape cash transfer policy and design as well as rhetoric, despite the absence of robust empirical evidence to justify these expectations.



21

Graduation

When targeted at potentially productive households, such as those with available labour, appropriately designed cash transfer programmes have the potential to enable households to gain access to, and benefit from, a range of other complementary development interventions, (such as programmes to promote agricultural productivity, microfinance, basic service provision, or the provision of a large cash transfer for asset purchase, together with a supporting stipend) and may in some instances result in graduation. In Ethiopia, households participating in the Productive Safety Net Programme (PSNP) make greater use of complementary credit facilities, and the health, education and water services produced under the public works component of the PSNP (Slater et al., 2006), which they can access due to receipt of the cash transfer. However to have a positive impact on employment prospects and wage income, such that beneficiaries can ‘graduate’ out of poverty, and out of the programme, cash transfers need to increase household asset portfolios, agricultural production and/or human capital development significantly. Currently there is little evidence internationally that this has happened, and often expectations of graduation are not proportionate to the scale of cash transfer programming, the value of the transfers themselves, or the extent of complementary developmental initiatives. The available evidence suggests that where cash transfer programmes are implemented in isolation from other complementary interventions to promote livelihoods, graduation is unlikely (ibid).

Growth

Similarly the impact of cash transfers on growth is often overstated (Slater, 2009). Even if it is assumed that cash transfer programmes result in improvements in household asset portfolios and productivity at a microeconomic level, there is little evidence on the impact of cash transfers on the local economy, or of cash transfer-induced increases in household welfare having a significant impact on national GDP in low income countries, largely due to the small scale of most programmes. This is not to say that cash transfers do not, or cannot, contribute to aggregate economic growth, if correctly designed and implemented as part of an integrated package of interventions, but the evidence is lacking (Barrientos, 2008). Cash transfers can affect local markets, by generating increased demand, which can, in turn, trigger a supply response by local producers. Conversely, where markets are not able to respond by increasing supply in this way, cash transfers can have a negative impact on the local economy by pushing up local prices. In Ethiopia, evidence from the Meket Livelihoods Programme demonstrates that shifting from food to cash-based transfer programmes had negative implications for the availability and price of food in local markets, especially in remote, food deficit areas, undermining prospects for both graduation and growth (Kebede, 2006).

It is the limited scale of most programmes; low levels of coverage, the low absolute value of many transfers, and the fact that cash transfer programmes tend to be implemented among economically marginalised populations, which limit the potential of most cash transfer programmes in terms of any significant macro-economic impact. These constraints are often overlooked in programme rhetoric, and expectations among both donors and national governments of cash transfer programming are often exaggerated, possibly in order to stimulate national government engagement, rather than focusing on the more mundane, but critical micro-economic impacts of cash transfer programming within a household, and the likely impact on the local community.



22

Implications for Effective European Union

Programming

The overview offered in this paper enables some conclusions to be drawn in terms of how donors may be able to enhance the potential for successful social protection programming in Low Income Countries by addressing programme integration, programme design and implementation, and most importantly financing modalities.

Integration

The integration and coordination of cash transfer programming with other development activities (agricultural extension, provision of quality services, infrastructure, micro-finance etc) can increase the likelihood of accumulation, productivity increases (and potentially graduation and growth).

For growth to take place, cash transfer programmes must be implemented on a large scale, with adequate national, district and community level coordination. A further requirement is the existence of capacity to extend the provision of other complementary programmes and services (such as health, education and agricultural extension). Such capacity is rare in low income countries, given the limited administrative, managerial and financial resources available, and coordination problems abound.

Ghana’s, Livelihood Empowerment Against Poverty (LEAP), illustrates the challenges of effective coordination across programmes. LEAP aims to link beneficiaries with complementary interventions, such as agricultural services and health care and insurance. The programme is still small-scale (currently in its pilot phase targeting those caring for orphans and vulnerable children) and experiencing a number of challenges in terms of effective delivery and the implementation of complementary services. The mapping of the availability of these additional services at the district or community levels has been inadequate, and the micro-credit and entrepreneurial programmes that were to be integrated with the cash transfer programme have been small-scale and not widespread, with limited additional financing for the provision of these services in beneficiary communities (Jones et al, 2008). Part of the reason for this is the limited institutional capacity to carry out practical coordination across multiple government agencies. Intra- and inter-agency information sharing has been weak and few government officers are well informed about LEAP implementation plans (ibid).

In Latin America, in contrast, cash transfer programmes tend to be better integrated, with a greater capacity for providing complementary services. In Mexico, for example, the Oportunidades programme administers income and nutritional supplements to five million households as well as providing conditional cash transfers. In LICs however, the lack of institutional capacity can result in trade-offs, rather than complementarity, at the operational level.

Donors can play a key role in addressing this by working closely with governments and through government structures, rather than programming in parallel and by ensuring donor financed programmes are integrated with other development interventions, and that institutional capacity is adequately supported, in order to prevent programmes operating in competition for scarce administrative and managerial resources.



23

Programme Design

While complementary programming is critical if growth at local or macro-levels is anticipated, there are a number of key programme design factors which can affect the impact of the programme at household level, in terms of poverty reduction and the potential for graduation; there is a need for cash transfers to be predictable, of adequate value, and provided on an ongoing basis. Cash transfers must be delivered predictably as this enables poor households to plan and make the best use of their resources - when transfers are delayed, households may have to take credit and lose a proportion of their transfer in debt payment when it eventually arrives. Also critical is the value of the transfer and the duration of transfer receipt. Where the value of the transfer is low (representing a small proportion of the poverty gap), the transfers tend to be consumed, and only when consumption needs are satisfied do households invest in human capital (education and health) and only once these needs are satisfied does investment in livelihoods ensue (Devereux, 2000).

Donors tend to play a key role in determining each of these aspects of cash transfer design; the value of the cash transfer, programme duration, and also the regularity of resource flows. Often where a number of donor-supported programmes are implemented in a single context, they exhibit a significant range of transfer values, participation duration, and payment predictability, based on donor specific preferences, norms and practices. This results in an inconsistency across programmes which increases the administrative burden on government implementers (particularly when considered in relation to the associated diversity of donor management and reporting requirements), and further reduces the likelihood of government buy-in. Donors can address this by moving away from the current situation of conflicting programming preferences based on individual institutional mandates to acknowledge donor preferences and accountability, and, instead, attempt to co-ordinate and rationalise programme development and support, and adopt an empirical basis for key design choices such as priority target-groups, targeting mechanisms, transfer levels, programme duration, etc.

Financing Modalities

The adoption of financing agreements which provide a credible assurance of medium term donor financing commitment may serve to reduce government concerns regarding the future liabilities implied by the continuation and extension of existing pilots, and the implementation of large scale national programmes. This may reduce government perceptions of the fiscal risk of engaging in large scale programming. The adoption of instruments such as multiple year financing packages, ten year initiatives, and donor financing baskets for social protection may serve this function of signalling credible ongoing donor commitment and thereby reducing public financial management concerns, and removing one barrier to Ministry of Finance acceptance of social protection programming.

In addition, improved donor coordination, rationalization and harmonization could also lead to greater efficiencies, and reduced transaction cost of cash transfer programming for governments.



24

Conclusion

The 2008 Social Policy Framework for Africa (SPF) includes the goal of introducing cash transfer schemes throughout the region. However, this initiative has not been translated into enthusiasm for cash transfer programming across governments in the region, or for commensurate budgetary allocations. Research suggests that governments may tolerate donor-funded cash transfer programming, even if it does not reflect domestic priorities, in part due to power imbalances between governments and donors. This tolerance does not necessarily imply political ownership, endorsement, or any significant financial commitment on the part of government to ongoing programme implementation or extension of pilots, and there is little evidence that budgetary reallocation in favour of this sector is seen as desirable or feasible in many LICs, particularly when at the behest of the donor community. Several pilot cash transfer programmes, including those used as models for cash transfer advocacy across Africa, are funded primarily by donors, with limited plans for significant state financing or budgetary reallocation. Where donor financing is short term, governments are reluctant to take on the liabilities implied by either pilots or going to scale.

A brief overview of cash transfer programming in sub-Saharan Africa suggests that where donor financing commitments are secured in the medium term and programme design conforms to national social protection preferences, there may be a greater potential for the success of donor-led cash transfer programming, and that there is also potential for donor support to domestically initiated and owned cash transfer programming. In order to promote cash transfer programming in the region donors need to engage more closely with governments to support national priorities, harmonise and rationalise their own programming, provide credible medium-term financing packages, ensure programme design is empirically, rather than institutionally, directed, and, for maximum impact, ensure that cash-transfer programming is integrated with other forms of development policy and implementation. If the European Union is able to adopt this approach in its own cash-transfer programming and to learn from the last decade of bi-lateral and multi-lateral cash transfer programming in the region, it will be well placed to make a significant contribution in this area in the future.



25

References

Barrientos, A. (2008) ‘Social Transfers and Growth Linkages’, Brooks World Poverty Institute/ University of Manchester, Manchester.

Bastagli, F. (2010) ‘From Social Safety Net to Social Policy? The Role of Conditional Cash Transfers in Welfare State Development in Latin America’. IPC Working Paper 60. International Policy Centre for Inclusive Growth, (IPC - IG), Bureau for Development Policy, UNDP. Brasilia, December.

Bastagli, F. (2010) ‘Poverty, Inequality and Public Cash Transfers: Lessons from Latin America’. Paper presented at the European Report on Development (ERD) conference on ‘Experiences and Lessons from Social Protection Programmes across the Developing World: What Role for the EU? Paris, 17-18 June 2010.

Chinsinga, B. (2009) ‘The Political Economy of Cash Transfers in Malawi’. A report prepared for ODI. Chancellor College.

Devereux, S. (2000) ‘Social Safety Nets for Poverty Alleviation in Southern Africa’. Brighton: Institute of Development Studies, University of Sussex.

Devereux, S. and White, P. 2010. Social|Protection in Africa: Evidence, Politics, and Rights. Poverty and Public Policy, Vol 2, issue 3,: 53-77.

Devereux, S., Davies, M., McCord, A., Slater, R., Freeland, N., Ellis, F., White, P and Sabates-Wheeler, R. (2010) Social protection in Africa: Where next? ODI, IDS, University of East Anglia and RHVP.

Du Toit, A. (2004) ‘Forgotten by the Highway: Globalisation, Adverse Incorporation and Chronic Poverty in a Commercial Farming District in South Africa’.Chronic. Poverty Res7earch Centre Working Paper No. 49. Cape Town and Manchester: Centre for Social Science Research and Chronic Poverty Research Centre.

Ellis, F. (2008) ‘“We are all poor here”: Economic difference, social divisiveness and targeting cash transfers in sub-Saharan Africa’. Paper presented at the conference Social Protection for the poorest in Africa: Learning from experience, Entebbe, Uganda, 8-10 September.

Habasonda, L. (2009) ‘Zambia, Political Economy of Cash Transfers’. A report prepared for ODI.

Handa, S. (2008) ‘The Orphan Targeting Dilemma in Eastern and Southern Africa’, pp18-19 in Cash Transfers: Lessons from Africa and Latin America. Poverty in Focus, International Poverty Centre August 2008.

Hanlon, Barrientos and Hulme. (2010) ‘Just Give Money to the Poor: The Development Revolution from the Global South’. Kumarian Press. Sterling. US.

Ikiara, G. K (2009) ‘The Political Economy of Cash Transfers in Kenya’. A report prepared for ODI. University of Nairobi.

ILO (2008) ‘Can low-income countries afford basic social security?’ Social security policy briefings, Paper 3. Geneva: ILO, Social Security Department.

Jones, N., Ahadzi, W. and D. Doh (2009) ‘Social protection and children in West and Central Africa: Case study Ghana’. ODI/UNICEF West and Central Africa Office (WCARO).

Kebede, E. (2006) ‘Moving from Emergency Food Aid to Predictable Cash Transfers: Recent Experience in Ethiopia’, Development Policy Review 24 (5): 579-599.

McCord, A. (2009) ‘Cash Transfers: Affordability and Sustainability’. Overseas Development Institute Project Briefing 30. ODI, London.

Slater, R. (2009) ‘Cash Transfers: Graduation and Growth’. Overseas Development Institute Project Briefing 29. ODI, London.



26

Slater, R. Ashley S. Tefera M. Buta, M. and Esubalew, D. (2006) ‘PSNP Policy, Programme and Institutional Linkages’. ODI, IDL Group Ltd. and Indak.

Slater, R. and Farrington, J. (2009) ‘Cash Transfers: Targeting’. Overseas Development Institute Project Briefing 27. ODI, London.

Wood, G. (1999) ‘Adverse Incorporation: Another Dark Side of Social Capital’. Working Paper for the Global Social Policy Project, University of Bath, UK. Mimeo 14 pages.

differing government and donor perspectives on …erd.eui.eu/media/backgroundpapers/mccord-differing...

Documents